Research On Sorting Method Of Frequency Hopping Communication Network Station

Frequency hopping signal has the advantages of good concealment and strong anti-interference.It has been widely used in military field.The accurate analysis of frequency hopping signal is the basis of its efficient countermeasure.The accurate estimation of its parameters and the efficient sorting of network stations are the two major problems faced in the accurate analysis of frequency hopping signal.This thesis focuses on the parameter estimation algorithm of frequency hopping communication system and the intelligent network sorting method.The main work is as follows:(1)A hop-timing estimation algorithm based on elastic network model and generalized alternating direction method of multipliers(ADMM)is proposed.Aiming at the problem that the parameter estimation accuracy of frequency hopping signal is not high,this method combines the sparse reconstruction algorithm based on elastic network model with the iterative idea of generalized ADMM method,and realizes the fast and accurate estimation of frequency hopping time of frequency hopping signal under the condition that the received signal frequency matches the preset frequency matrix.Simulation results show that when the number of iterations is reduced by 60%,the proposed algorithm can obtain a clear time-frequency pattern under the condition of low signal-to-noise ratio,and the accuracy of hop-timing estimation is greatly improved.(2)A hop-timing estimation algorithm based on frequency matrix correction is proposed.This method constructs the frequency matrix correction model based on Taylor series expansion,iteratively solves the frequency deviation term and updates the frequency matrix through the expectation maximization algorithm,and completes the accurate estimation of the hop-timing based on the updated frequency matrix.It solves the problem that the reconstructed time-frequency matrix is inaccurate due to the mismatch between the received signal frequency and the preset frequency matrix,which affects the hop-timing estimation.Simulation results show that the proposed algorithm can effectively correct the frequency matrix,and the accuracy of hop-timing estimation based on the corrected frequency matrix is greatly improved.(3)A direction of arrival(DOA)estimation algorithm based on single source point detection is proposed.Firstly,the time-frequency points dominated by a single signal source are detected based on the single source point detection criterion and the interference points are removed.Then the mixing matrix is estimated by clustering them based on nonparametric Bayesian model.Finally,the subspace projection algorithm is used to recover the source signal and obtain DOA information.Simulation results show that the proposed algorithm can greatly improve the accuracy of DOA estimation under the condition of low signal-to-noise ratio without setting the number of clusters in advance.(4)The scheme of frequency hopping network station sorting is designed.Based on the estimation of frequency hopping signal parameters,the scheme completes the sorting of frequency hopping stations by optimizing the improved k-means algorithm for the selection of initial clustering centers.For the asynchronous frequency hopping signal,the asynchronous frequency hopping network station sorting is completed by using the hopping time information.For synchronous frequency hopping signals,DOA information is used to complete the sorting of synchronous frequency hopping stations.Based on the measured data,the verification results show that when the signal-to-noise ratio is greater than 5d B,the sorting accuracy of this scheme exceeds 97%.